A Self-Health-Learning GaN Power Converter Using OnDie Logarithm-Based Analog SGD Supervised Learning and Online TJ-Independent Precursor Measurement

📋 论文概要

该论文提出一种自健康学习的GaN功率转换器，利用片上基于对数的模拟随机梯度下降（SGD）监督学习算法和在线结温监测，解决GaN功率电路的可靠性问题。通过实时健康监测和自适应学习，提升GaN功率转换器的寿命和稳定性。

💡 主要创新点

• 首次将基于对数的模拟SGD监督学习集成到GaN功率转换器中，实现片上自适应健康管理。
• 提出在线结温（TJ）监测技术，结合自学习算法动态优化功率器件工作状态。

🏷 关键词

📄 原文摘要

As GaN technology proliferates in modern power electronics, reliability of GaNbased circuits has become the biggest hurdle for commercialization. Sustaining largest voltage and current stresses in power circuits, power devices on average account for over 31% of failures [1]. With new problems such as current collapse and thermal aging, GaN power circuits deem to face more reliability challenges compared to their silicon counterparts [2]. In such a situation, health condition monitoring is of paramount importance. As shown in Fig. 18.1.1, due to hot electron injection and charge trapping effects, current collapse weakens 2dimensional electron gas (2DEG) layer in a GaN switch over time, elevating its dynamic on-resistance rDS_ON gradually. The clear link between rDS_ON and aging (Fig." 18.1.1) makes rDS_ON a widely accepted precursor for GaN condition monitoring [3-5]. However, measuring rDS_ON is not a simple task. Traditionally,

👥 作者与机构

Yuanqing Huang, Yingping Chen, D. Brian Ma

The University of Texas at Dallas, Richardson, TX